FREE AND OPEN SOURCE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

A Survey On Embedded Open Source System Software For The

Mahdi Amiri-Kordestani1 Hadj Bourdoucen1,2 1 Communication and Information Research Center, 1 Communication and Information Research Center, Sultan Qaboos University, 2 College of Engineering, ECE Dept. Muscat, Oman Sultan Qaboos University, [email protected] Muscat, Oman, [email protected]

Abstract— Internet of Things (IoT) is an opportunity for and very similar to open source software so as, anyone can future devices to be smarter, more robust and efficient. This examine, change, distribute, produce, and market devices opening has become available due to the continuous cost based on that design. Open source hardware also provides reductions of many individual systems and components such the freedom to utilize and advance the device while sharing as sensors, computing devices, communications methods, the experiences. cloud and the big data paradigms. Open source software, hardware, and standards are considered as further cost-saving There are hundreds of open source projects about the IoT ways, and thus, a significant number of current products are and currently looking at .com with phrase “Internet of already using open source projects as their principal elements Things” represents more than 1,800 projects. Major IoT to reduce the development time, minimize the cost, and projects and frameworks usually address a typical ecosystem develop more competitive products. that lets IoT devices communicate to the Internet, and However, choosing the right components for the open usually to a cloud system that can store and process the data source system depends on multiple factors, such as the open and generate advanced reports and controlling signals [6]. source licensing, application complexity, required performance, hardware architecture, communication Based on mentioned pattern, a considerable number of standards, external sensors, peripherals, and other open source projects are available that address one or more parameters. This paper studies a number of the state-of-the- of the following concepts: art open source system software projects and frameworks and identifies some of the key parameters that should be 1. software for IoT devices. considered to select and adapt for an open source project. 2. Network protocols such as LE, ZigBee, Moreover, a comparison of some of the favorite software, , and 6LoWPAN techniques. hardware and standards that target the Internet of Things is 3. Application layer protocols such as CoAP and represented. MQTT. 4. Security and privacy mechanism such as M2M Keywords—Internet of Things; System Software; authentication and cryptography (AES, SSL/TLS). Embedded ; Open Source 5. Higher level management, control, and reporting. I. INTRODUCTION 6. Application and firmware upgrade and patch management. Nearly all the leading technology and business forecasts agree that Internet of Things (IoT) is going to be a huge II. IOT APPLICATION SPECTRUM market and will cause a significant impact on our lives [1]– The "Internet of Things" is a comprehensive model of all [3]. IoT is a combination of many components with different kinds of computing devices, (also referred to "things," scopes that includes a variety of sensors, actuators, "embedded devices," and "smart devices") that are somehow embedded computing systems, communications techniques, connected to the Internet. In theory, a device can be attached as well as various protocols, frameworks, middleware, and to almost any object such as vehicles, home appliances, software systems. The IoT is frequently associated with industrial mechanical or electrical machines, and even a cloud computing and the big data techniques as the primary person to let the object to communicate to the Internet. data end points. Some major IoT applications [7] are buildings and home Open source software as specified by the Free Software automation, smart cities, smart industry or manufacturing, Foundation (FSF) [4], allows the user to run, share, study, wearables and healthcare devices, as well as automotive. and modify the software. In other words, the term free here TABLE 1, indicates a number of major IoT fields, a typical means freedom in the following abilities: application, and some average communication distance  Use the software for any purpose. ranges.  Copy and distribute the software to other people. TABLE 1. TYPICAL IOT APPLICATIONS AND THEIR RANGE  Study the algorithm and learn the technology Field Application Range  Develop and improve the software. Wearables Healthcare, WBAN <2m Automotive Infotainment <5m From the hardware aspect, “Open Source Hardware Smart homes Intelligent buildings <100m Association” (OSHWA) [5] describes the open source Smart industries Smart control <10 Km Smart cities Smart traffic control <100 Km hardware as the hardware with a publicly available design Smart grid Power distribution 100+ Km

© SQU-2017 ISSN: 1813-419X -27-

FREE AND OPEN SOURCE SOFTWARE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

A major market for IoT applications is the home systems, home monitoring, and security systems as well as appliances that make smart home [8] to be the most popular academic research titles such as intelligent homes and IoT application among other fields. The smart home includes buildings to reduce the overall consumed energy. many available commercial products such home automation

TABLE 2. COMPARISON BETWEEN SOME MAJOR IOT CONNECTIVITY STANDARDS Standard Native IP 2.4 GHz Mesh Approximate Range Data Rate Example Application Wifi 802.11 b/g/n Yes Yes No a 35-250 m 600 Mbps (802.11n) Home, Industry Bluetooth Smart (LE) No Yes No 50-100 m 1 Mbps Home, Wearables ANT+ No Yes Yes 30 m 12.8/60 Kbps Home, Wearables ZigBee (802.15.4) No Yes b Yes 20 m 20 to 250 kbps Home, Industry Thread (802.15.4) Yes Yes Yes 20 m 250 kbps Home, Industry Z-Wave No No Yes 30 m 9.6/40/100 Kbps Home, Industry Sub-Ghz n/a c No n/a c 100+ Km n/a c , Grid Cellular Yes (4G) No No 35 Km (GSM), 200 Km (HSPA) 50-100 Mbps (LTE) Smart City, Grid a. 802.11s also adds the support for mesh networks. b. ZigBee can operate at 2.4 GHz as well as 868 MHz in EU and 915 MHz in US. c. Z-Wave operates at 868.42 MHz in EU and 908.42 MHz in US, other countries vary. d. Sub-Ghz specification depends on the implementation.

Wearable [9] and healthcare devices are the second most standards as represented in TABLE 2, with different popular application of the IoT. A wearable is a small hardware architectures as illustrated in TABLE 3 and computing device running on a battery that usually includes consequently, there are many system software and projects some microelectromechanical sensors (MEMS) to monitor for various needs. The system software which is also called vital signs such as motions, temperature, heart rate, blood operating system for larger computing devices and firmware pressure, and others. Some typical wearable applications are for smaller embedded systems is usually a type of program for entertainment, fitness, smart watches, as well as that is designed to run on hardware directly. A number of healthcare and monitoring devices. As an example, a open source system software is discussed here, and a healthcare device can monitor vital signs and transfer the comparison is shown in TABLE 4. captured data to an upper system such on the Internet (probably a cloud system) through network gateways or A. Collaborative projects other communication mechanisms such as cellular networks. Many important open source projects follow the type of The health monitoring device and the management system collaborative projects and are governed by a well-known can alert both the person as well as corresponding doctors in organization such as the Linux Foundation and a list of case of difficulties and emergencies. corporate members from various industries. Linux Foundation as a key organization in open source is a non- The solutions such as smart cities [8] expand the IoT profit organization with an aim to promote, support and domain to a wider area. Some major topics are about greater improve Linux and some key open source projects. management of traffics, water, waste, and the city security. Currently, there are more than 50 projects listed on the Some specialized topics such as smart grids [10] expand the organization’s website, and some of those projects are IoT to a much wider and to a country or state. The idea directly targeting the IoT domain such as IoTivity, Tizen, behind smart grids is to monitor both sides of electricity Automotive Grade Linux, Yocto, and Zephyr. consumers and suppliers to optimize the power distribution and increase the efficiency and safety. IoTivity [11] is an open source framework of the connectivity standards for IoT that was originally developed IoT applications in industry such as smart industry, smart by to allow devices to communicate to the Internet and manufacturing or industrial Internet [7], introduces the to each other. Another very similar open source project was combination of electrical and mechanical machines, the AllJoyn framework by Allseen Alliance which is advanced data analytics and people at work to increase the currently merging with IoTivity in a unified framework. overall performance and productions. The combination of Organizations such as Linux Foundation, Open Connectivity devices with network sensors and artificial intelligence can Foundation (OCF), AllSeen Alliance and many major better serve the industries to gain productions. companies such as Intel, Microsoft, Samsung, LG, Another trending application for IoT is automotive , Mediatek, CISCO, GE, Canon, Sharp and others industry [7]. Some common patterns are vehicle to vehicle are all involved in this huge project. (V2V) communications, vehicle to roadside Tizen [12] is a Linux based distribution and a competitor communications, and the network inside vehicles that links to both ’s Android and Apple’s iOS that can address all sensors, cameras, the navigation system, and other a broad range of embedded devices, including smartphones, embedded devices to communicate and collaborate with tablets, TVs, players, cameras, printers, home appliances, each other. and wearables. Samsung is the major player of Tizen project and has merged its Bada project into Tizen. The other III. OPEN SOURCE IOT SOFTWARE leading members from silicon and telecom industries are As discussed above, the IoT applications have a broad Intel, NEC, Fujitsu, , Panasonic, Sprint, range of domains, and they incorporate many network and others.

© SQU-2017 ISSN: 1813-419X -28-

FREE AND OPEN SOURCE SOFTWARE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

Automotive Grade Linux (AGL) [13] is an embedded River Linux are based on specific builds of Yocto Linux Linux project to create open source applications and project. solutions with members from the automotive, telecom, and silicon industries. The primary purpose for AGL is In- Zephyr [15] is somehow a very different developing Vehicle-Infotainment (IVI) systems, but further applications project from Linux Foundation that provides an embedded such as digital instrument panels and telematic systems are and real-time operating system (RTOS) for also planned to be added. There is also an AGL Linux with concentrated resources that cannot use the standard distribution that collects some of the software that has been Linux kernel or they require specialized features such as originally developed for other platforms such as Tizen IVI real-time or safety specifications. Intel (Wind River and GENIVI in a Linux distribution. Systems) originally developed Rocket kernel for embedded and IoT devices. Later, Intel donated the Rocket kernel to The Yocto [14] is an embedded Linux project that Zephyr project to cooperate with other members. Zephyr provides a framework to create a highly customized Linux as illustrated in TABLE 4 has a more distribution. Some well-known embedded and commercial permissive license compared to other open source RTOS grade projects such as Mentor Embedded Linux and Wind projects such as FreeRTOS as well as the original Linux kernel source code.

TABLE 3. ARM ARCHITECTURES AND EXAMPLE OF IOT APPLICATIONS. Architecture Bits ARM Profile Suitable IoT devices Common cores New cores ARMv8-A 32/64 Application Ultra-high-performance, full OS Cortex-A53, A57, A72 Cortex-A35, A73 ARMv8-A 32 Application High-performance, full OS New Cortex-A32 ARMv8-R 32 Real-time Real-time, High-performance, RTOS New Cortex-R52 ARMv8-M 32 Controller High-performance, RTOS New Cortex-M23, M33 ARMv7-A 32 Application High-performance, full OS Cortex-A5, A7, A8, A9, A15 Cortex-A17 ARMv7-R 32 Real-time Real-time, RTOS Cortex-R4, R5, R7 Cortex-R8 ARMv7E-M 32 Controller Low-power, RTOS plus DSP Cortex-M4, M7 n/a ARMv7-M 32 Controller Low-power, RTOS Cortex-M3 n/a ARMv6-M 32 Controller Ultra Low-power, RTOS Cortex-M0, M0+, M1 n/a (Full OS stands for full feature operating systems such as Linux, RTOS stands for bare-metal applications or real-time operating systems)

B. Android ecosystem and power compared to AOSP. The Android Things also has Google is a key company regarding the IoT field because Android ecosystem and utilizes the existing development it is directing the Android software as the base software for tools, APIs, and resources that also provide low-level I/O many smartphones. Meanwhile, it has the infrastructure and controls and required libraries for some sensors, display the required experiences for running big data models and controllers, and embedded modules. cloud services. Currently, Google is offering two products C. ARM ecosystem that target the IoT, the Weave [16] as a communications platform and Android Things (formerly Brillo) [17] as an ARM ecosystem is mostly focused on the system Android-based operating system as well as necessary software for low-cost and low-power Cortex-M processors. frameworks for high-end products very similar to Android There are many types of hardware devices, and hence there operating system for smartphones. are also many different system software products. For instance, in ARM Cortex processor family, there are three Google Weave is a competitor of IoTivity framework profiles of processors to address various applications as and provides a communication and controlling mechanism ARM Cortex-A, Cortex-R, and Cortex-M. based on HTTPS for controllable devices such as lights and wall switches. Weave has two components; "Weave Device ARM Cortex-A family as represented in TABLE 3, is a SDK" and "Weave Server." The device SDK is an open full-featured application processor with many components such as memory management unit (MMU) and is usually source library (libiota) written in C language for maximum portability and can be integrated with IoT devices to let them used in high-end devices such as smartphones, tablets, and communicate with the Weave servers. Currently, the SDK netbooks. Cortex-A can handle large applications and full can only support HVAC controller, light, outlet, television, feature operating systems such as Android, Linux, and and wall switch as IoT device types, and there is a promise Windows. ARM Cortex-R is designed to be a deterministic to add other kinds to this project. The other component of processor and to address real-time and functional safety the Weave platform is the server side "Weave Server" that applications. The Cortex-M is the smallest member of the presents secure methods to store the states, device family with much lower cost, performance and power registration mechanism, and command broadcasting. The consumption that is more suitable for many IoT devices. As server is already integrated with other Google cloud an example, an ultra-low power ARM Cortex-M0+ can run services. off a battery for years [18], and is still powerful enough for many IoT devices. Android Things follows Android Open Source Project (AOSP) patterns and software that is designed to support As discussed earlier, Android Things (Brillo) is suitable high-end IoT devices, sensors, and wireless networking for high-performance processors such as ARM Cortex-A, while claiming much lower amounts of storage, memory, and architectures. Meanwhile, the Zephyr project from

© SQU-2017 ISSN: 1813-419X -29-

FREE AND OPEN SOURCE SOFTWARE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

Linux Foundation is designed for low-performance RIOT [21], [22] is another open source (LGPL) and controllers such as Cortex-M that lack features such as partially POSIX compatible system software with real-time MMU. support that targets a broad range of 8-bit, 16-bit and 32-bit microcontrollers. RIOT is designed to be energy- mbed is the official ARM solution designed to address efficient and modular with very low memory requirements. the embedded devices and the Internet of Things. The "mbed RIOT was initially developed as a research project by FU OS" and ARM's upcoming product "mbed Cloud" are the Berlin, INRIA, and HAW Hamburg and follows the two main components of this project [19]. mbed OS is an microkernel type, somehow inherited from FireKernel [23] open source system software (Apache 2.0 license) that project that was targeting wireless sensor networks. RIOT targets Cortex-M microcontrollers, and incorporates a tiny supports the essential features such as memory allocations RTOS based on CMSIS-RTOS RTX, security features and multi-threading for C and C++ programming languages (TLS), many connectivity standards, and device drivers for as an advantage. RIOT already includes some of common popular sensors in a modular form. mbed OS supports an networking protocols and standards such as IPv6, extensive range of communication standards such as 6LoWPAN, UDP, RPL, CoAP, and CCN-Lite. Bluetooth LE, 6LoWPAN, Ethernet, WiFi, RFID, Cellular, Thread, and LoRa LPWAN. mbed OS is designed to be [24] is a new developing system compatible with upcoming mbed Cloud product from ARM software project and is a real-time and has a modular and that will add some cloud features such as mbed Cloud open-source architecture designed for low-performance IoT Provision, Connect, and Update services to mbed ecosystem. devices. Mynewt targets ARM Cortex M0 to 4 and RISC-V architectures with a plan to extend the hardware support to D. Community system software projects MIPS architecture. Mynewt targets low-cost devices that Other than mentioned extensive ecosystems, many have to operate for an extended period with limited power, individual projects have turned into key components for IoT memory, and flash storages. It is a collaborative open-source devices. [20] is a well-known system software for project with Apache License 2.0 governed by the Apache both academic research and industrial usage that initially is Software Foundation. Mynewt includes NimBLE [25] as the designed to target the low-power and low-cost first fully open source implementation of Bluetooth Low microcontrollers such as TI MSP430x, AVR, and Energy (BLE) 4.2 as an advantage to the other projects. newer TI SimpleLink family. Contiki and its native simulator Cooja, have been used as research vehicles for There are also many other Linux-based embedded many projects. Contiki is best distributions such as Ubuntu Snappy Core, and Ostro that try known for the complete and efficient implementation of to address high-performance devices. These are not covered essential network standards such as IP, UDP, TCP, as well in this paper. as low-power standards such as 6LoWPAN, RPL, and CoAP.

TABLE 4. COMPARISON BETWEEN ARM MBED, ANDROID THINGS, AND SOME COMMUNITY PROJECT. System Software Provider License Example Hardware Integrated Cloud High-end processors such as Android Things Google Apache 2.0/GPLv2 Google Weave x86, ARM Cortex-A (32/64) mbed OS ARM Apache 2.0 ARM Cortex-M mbed Cloud Community; ARM Cortex-M, MSP430, Contiki OS 3-clause BSD Thingsquare Thingsquare AVR, x86 Community; ARM Cortex-M, MSP430, RIOT OS LGPLv2.1 n/a FU Berlin ARM7, AVR, x86 Collaborative by ARM Cortex-M, x86, Zephyr Apache 2.0 n/a Linux Foundation ARC, NIOS2 Collaborative by Apache Mynewt Apache 2.0 Cortex-M, RISC-V n/a Apache Foundation

IV. RESEARCH OPPORTUNITIES hardware interfaces and resources, and even backdoors and malware created during the development or manufacturing. Communication standards, security, and privacy are a number of significant research areas for IoT devices [2], Recent research on some IoT devices [28] reports that [26], [27] because the connected things are not always 90% of embedded devices including their ecosystem general purpose computers, but a variety of embedded software collect some personal information and 70% of them devices that can communicate to each other and to the communicate through unencrypted network services. About, Internet. For instance, an insecure device can compromise 60% of the system software of investigated devices could be and allow unauthorized users to gain necessary to upgraded through the network without using any data perform the monitoring or controlling the device or a system encryption or protection mechanisms. Moreover, some and collect private data or disturb the owner. update images (firmware and patches) where standard Linux images could easily be mounted and altered by Security risks for IoT devices can be caused by different ordinary users. reasons [27] such as the lack of experience regarding security concerns, the absence of aftersales bug fixes or updates, problems during software upgrades, constrained

© SQU-2017 ISSN: 1813-419X -30-

FREE AND OPEN SOURCE SOFTWARE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

V. CONCLUSIONS REFERENCES The Internet of Things will bring shortly many [1] J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, “Internet embedded smart things to our life. Many open source of Things (IoT): A vision, architectural elements, and future components such as system software projects can be directions,” Futur. Gener. Comput. Syst., vol. 29, no. 7, pp. employed to accelerate the development speed and reduce 1645–1660, 2013. the overall cost of future devices. [2] L. Atzori, A. Iera, and G. Morabito, “The Internet of Things: A Some primary concerns in choosing an open source survey,” Comput. Networks, vol. 54, no. 15, pp. 2787–2805, system software other than technical specifications (such as 2010. support for hardware, functionality, reliability, performance, network connectivity and standards) are the type of software [3] J. Manyika et al., “The Internet of Things: Mapping the value license, project governance, founding members, commercial beyond the hype.,” McKinsey Global Institute, 2015. [Online]. and community supports, and finally the size of users. Available: http://www.mckinsey.com/business-functions/digital- mckinsey/our-insights/the-internet-of-things-the-value-of- A number of major ecosystems have been introduced and digitizing-the-physical-world. [Accessed: 03-Jan-2017]. studied in this paper. Among all those projects, the IoTivity from Linux Foundation has an active and massive [4] “What is free software?,” Free Software Foundation. [Online]. community from reputable companies with a permissive Available: https://www.gnu.org/philosophy/free-sw.html. license without commercial components. [Accessed: 24-Dec-2016]. Google Android Things in combination with Android [5] “Open Source Hardware (OSHW) Definition,” Open Source operating system, Google cloud infrastructure, and other Hardware Association. [Online]. Available: cloud services is also a turnkey solution that will empower http://www.oshwa.org/definition/. [Accessed: 24-Dec-2016]. lots of high-performance devices in the near future but will [6] H. Derhamy, J. Eliasson, J. Delsing, and P. Priller, “A survey of lock solution providers to Google ecosystem. commercial frameworks for the Internet of Things,” in 2015 The other ecosystem is ARM mbed OS in combination IEEE 20th Conference on Emerging Technologies Factory with mbed Cloud as a chip to the cloud solution that can Automation (ETFA), 2015, pp. 1–8. target low-performance MCU devices but will also lock [7] Internet of Things Applications - From Research and Innovation solution providers to mbed Cloud ecosystem. to Market Deployment (River Publishers Series in Contiki, Zephyr, Mynewt, and RIOT introduced in this Communications). River Publishers, 2014. paper (summarized in TABLE 4) are open source system [8] D. Miorandi, S. Sicari, F. De Pellegrini, and I. Chlamtac, software without integrated commercial cloud services that “Internet of things: Vision, applications and research can be utilized as essential components for larger ecosystems challenges,” Ad Hoc Networks, vol. 10, no. 7, pp. 1497–1516, systems. 2012. Contiki and RIOT are both mature system software [9] A. Pantelopoulos and N. G. Bourbakis, “A Survey on Wearable projects for IoT devices. Contiki has an extensive Sensor-Based Systems for Health Monitoring and Prognosis,” networking stack and a strong community that is supported IEEE Trans. Syst. Man, Cybern. Part C (Applications Rev., vol. by a number of large companies. There is also an integrated 40, no. 1, pp. 1–12, Jan. 2010. commercial cloud solution from the Thingsquare that extends and connects this project to the cloud. It is worth [10] V. C. Gungor et al., “Smart Grid Technologies: Communication noting that RIOT better supports high-level languages such Technologies and Standards,” IEEE Trans. Ind. Informatics, vol. as C and C++ and has a better memory management with a 7, no. 4, pp. 529–539, Nov. 2011. multi-thread support but offers a restricted “LGPLv2.1” [11] “IoTivity project,” Linux Foundation. [Online]. Available: compared to the “3-clause BSD” license https://www.iotivity.org/about. [Accessed: 24-Dec-2016]. used by Contiki OS. [12] “TIZEN,” Linux Foundation. [Online]. Available: Zephyr and Mynewt are two emerging collaborative https://www.tizen.org/about. [Accessed: 24-Dec-2016]. projects from Linux Foundation and Apache Foundation [13] “Automotive Grade Linux,” Linux Foundation. [Online]. with Apache 2.0 as a permissive open source license. Both Available: https://www.automotivelinux.org/about. [Accessed: projects are quickly growing by adding hardware supports, software, and connectivity features and they will capture 24-Dec-2016]. more interest in the future from IoT vendors. [14] “,” Linux Foundation. [Online]. Available: https://www.yoctoproject.org/about. [Accessed: 24-Dec-2016]. ACKNOWLEDGMENT [15] “Zephyr Project,” Linux Foundation. [Online]. Available: This work is part of the research that is conducted at the https://www.zephyrproject.org/about. “Free and Open Source Software Laboratory” at [16] “Google Weave,” Google. [Online]. Available: Communication and Information Research Center (CIRC), https://developers.google.com/weave/. [Accessed: 24-Dec-2016]. Sultan Qaboos University, Muscat, Oman. [17] “Android Things,” Google. [Online]. Available: https://developer.android.com/things/index.html. [Accessed: 24- Dec-2016].

© SQU-2017 ISSN: 1813-419X -31-

FREE AND OPEN SOURCE SOFTWARE CONFERENCE (FOSSC-17) MUSCAT, FEBRUARY 14-15, 2017

[18] “Microchip-Atmel: SAM L MCUs,” Microchip - Atmel, 2017. 834–841. [Online]. Available: [24] “Apache Mynewt: An OS to build, deploy and securely manage http://www.atmel.com/products/microcontrollers/ARM/SAM- billions of devices,” Apache Software Foundation, 2017. L.aspx. [Accessed: 06-Jan-2017]. [Online]. Available: https://mynewt.apache.org/about/. [19] “ARM mbed,” ARM. [Online]. Available: [Accessed: 06-Jan-2017]. https://www.mbed.com/en/. [Accessed: 24-Dec-2016]. [25] “NimBLE Introduction,” Apache Software Foundation, 2017. [20] A. Dunkels, O. Schmidt, N. Finne, J. Eriksson, F. Österlind, and [Online]. Available: N. T. M. Durvy, “The Contiki OS: The Operating System for the http://mynewt.apache.org/network/ble/ble_intro/. [Accessed: 06- Internet of Things,” Contiki OS, 2011. [Online]. Available: Jan-2017]. http://contiki-os.org/. [Accessed: 24-Dec-2016]. [26] S. Li, L. Da Xu, and S. Zhao, “The internet of things: a survey,” [21] “RIOT: The friendly Operating System for the Internet of Inf. Syst. Front., vol. 17, no. 2, pp. 243–259, 2015. Things,” RIOT. [Online]. Available: http://www.riot-os.org/. [27] F. Baker et al., “Internet of Things (IoT) Security and Privacy [Accessed: 24-Dec-2016]. Recommendations,” Broadband Internet Technical Advisory [22] E. Baccelli, O. Hahm, M. Gunes, M. Wahlisch, and T. C. Group, 2016. [Online]. Available: https://www.bitag.org/report- Schmidt, “RIOT OS: Towards an OS for the Internet of Things,” internet-of-things-security-privacy-recommendations.php. in 2013 IEEE Conference on Computer Communications [28] “Internet of things research study,” Hewlett Packard Enterprise Workshops (INFOCOM WKSHPS), 2013, pp. 79–80. Development LP, 2015. [Online]. Available: [23] H. Will, K. Schleiser, and J. Schiller, “A real-time kernel for https://www.hpe.com/h20195/v2/GetPDF.aspx/4AA5- wireless sensor networks employed in rescue scenarios,” in 2009 4759ENN.. [Accessed: 01-Jan-2017]. IEEE 34th Conference on Local Computer Networks, 2009, pp.

© SQU-2017 ISSN: 1813-419X -32-